Functional electrical stimulation (“FES”) is the application of stimulation devices to nerves and muscles to treat medical disorders. The most successful FES system to date is the cardiac pacer which has become a routine part of cardiac disease therapy: Lynch, Cardiovascular Implants, in Implants, Lynch ed., Van Nostrand Rheinhold, New York 1982, incorporated herein by reference. However, there are a variety of other FES systems. The most heavily researched are FES systems to restore locomotion to paraplegics and arm motion to quadriplegics: Peckham, IEEE Trans. Biomed. Eng. 1991, 28: 530, incorporated herein by reference. Other motor control devices restore bladder control to paraplegics and diaphragm function to high quadriplegics: Erlandson, Scand. J. Urol. Nephrol. 44 Suppl: 31, 1978; Glenn, Ann. Surg. 183: 566, 1976, incorporated herein by reference. There are also FES devices designed to rehabilitate the sensory deficits, such as the cochlear implant: Hambrecht, Ann. Otol. Rhinol Laryngol. 88: 729, 1979, incorporated herein by reference.
The recurrent laryngeal nerve, which innervates the larynx, contains motor fibers that innervate both the abductor/opener and adductor/closer muscles of the vocal folds. Damage to this nerve compromises both of these functions and arrests the vocal fold just lateral to the midline. In unilateral paralysis, the voice is breathy and aspiration can occur because of compromised adduction, but airflow during inspiration is minimally impaired. Adequate ventilation of the lungs is assured because abduction of the opposite fold can still occur with each inspiration. In bilateral paralysis, there is a loss of abductory function in both folds, the voice may be minimally impaired because of fold symmetry and their paramedian position in most of the patients, but airway discomfiture is usually severe. Typically, the patient can tolerate restricted activity or may be relegated to a sedentary lifestyle until treatment is administered. In some situations, however, the condition may be life-threatening.
Clinical management of vocal fold paralysis focuses on the major laryngeal dysfunction associated with each of these two main types. Conventional treatments for unilateral paralysis aim at medializing the fold to improve voice production. Treatment for bilateral paralysis typically requires a tracheotomy to restore sufficient airflow to the lungs. The tracheotomy is left in place until nerve regeneration and muscle reinnervation has returned. However, in many cases, muscle reinnervation is either incomplete or inappropriate resulting in chronic paralysis. Under such conditions, surgical resection of the vocal fold (i.e., cordotomy) is employed to permanently increase the airway and relieve the patient of his tracheotomy. Although these conventional methods of treatment have been useful, they are less than ideal, since they tend to improve upon one laryngeal function at the expense of another. For example, cordotomy improves ventilation, but compromises voice production and airway protection.
Surgical techniques, such as laser arytenoidectomy and partial cordectomy, can be performed to widen the airway and relieve dyspnea in the case of chronic paralysis. However, these procedures compromise voice and airway protection to restore ventilation through the mouth. They also ignore the long-term effects of ensuing atrophy on vocal fold mass and position. In general, the greater the cartilaginous or membranous resection associated with either technique, the greater the morbidity. A number of modifications of these two strategies have been devised in an attempt to strike a more delicate balance between improved oral ventilation and impaired voice and swallowing. However, a more conservative stance toward resection increases the probability of failed intervention and the necessity for revision surgery. A new, more physiological approach termed laryngeal pacing has been studied in animal models as a means to restore oral ventilation.
Application of FES to paralyzed laryngeal muscles was introduced into human clinical otolaryngology in 1977 by Zealear D L, Dedo H H, Control Of Paralyzed Axial Muscles By Electrical Stimulation, Acta Otolaryngol (Stockholm) 1977, 83:514-27, incorporated herein by reference, which specifically addressed the case of unilateral vocal fold paralysis. Patients normally breathe well, but they cannot approximate both vocal folds. As a result, their voice is weak and breathy, and they tend to aspirate fluids. Zealear and Dedo proposed that a unilaterally paralyzed patient could be reanimated to close appropriately by electrical stimulation triggered by signals relayed from its contralateral partner. As simpler surgical methods were discovered to restore function in unilateral vocal fold paralysis, the development of an implantable neuroprostheses for this condition has not been vigorously pursued.
Mayr, Zrunek, et al., A Laryngeal Pacemaker For Inspiration Controlled Direct Electrical Stimulation Of Denervated Posterior Cricoarytaenoid Muscle In Sheep, Eur. Arch. Otorhinolaryngol, 248(8):445-448, 1991, incorporated herein by reference, described 8 sheep with denervated PCAs which received implants for from 5-18 months, and ruled out reinnervation by control.
Obert et al., Use Of Direct Posterior Cricoarytenoid Stimulation In Laryngeal Paralysis, Arch. Otolaryngol 1984, 110: 88-92, incorporated herein by reference, restored full abduction in bilaterally denervated dogs implanted with single-stranded teflon electrodes, using 20 ms stimulus pulses delivered at 20-40 Hz and 2-3 mA. Their study suggested that stimulus pulses should be synchronized with inspiratory signals in abductor pacing. Bergmann et al., Respiratory Rhythmically Regulated Electrical Stimulation Of Paralyzed Muscles, Laryngoscope, 1984, 94:1376-80, incorporated herein by reference, successfully implanted this idea of respiratory regulation of stimuli, using signals relayed from chest wall expansion. Canine PCA muscles were activated using parameters of 30 Hz, 1 ms, and large amplitudes of up to 50 mA.
Kano and Sasaki, Pacing Parameters of the Canine Posterior Cricoarytenoid Muscle, Ann. Otol. Rhinol Laryngol., 100:584-588, 1991, incorporated herein by reference, used a pair of coiled electrodes, separated by 2 mm, to stimulate the PCA. They observed promising abductions at 60-90 Hz and 2 ms. Bergmann et al reported 2-3 mm of abduction with stimulation of the PCA using a stimulus delivery system that had been chronically implanted for 11 months.
Otto et al, Coordinated Electrical Pacing Of Vocal Cord Abductors In Recurrent Laryngeal Nerve Paralysis, Otolaryngol. Head Neck Surg., 1985, 93:634-8, incorporated herein by reference, used electromyographic (EMG) signals from the diaphragm to regulate stimuli to denervated canine PCA muscles, and reportedly restored full abduction of the glottis.
Zealear and Herzon, Technical Approach For Reanimation Of The Chronically Denervated Larynx By Means Of Functional Electrical Stimulation, Ann. Otol. Rhinol Laryngol., 1994 September, 103(9):705-12, incorporated herein by reference, first introduced use of tiny coiled electrodes for abductor pacing in a study of inspiratory trigger sources including tracheal elongation, diaphragm EMG signals, phrenic nerve activity, and intrathoracic pressure changes.
Zealear et al, Technical Approach For Reanimation Of The Chronically Denervated Larynx By Means Of Functional Electrical Stimulation, Ann. Otol. Rhinol Laryngol. 1994, 103: 705-12, incorporated herein by reference, implanted an electrode array 3 months after RLN section, and the paralyzed stump was electro stimulated to rule out reinnervation. The hot spots were located in the middle of the PCA muscle, several millimeters from the median raphe, and covered 30-40% of the muscle surface area.
During chronic pacing, it would be desirable to stimulate above the fusion frequency for the PCA muscle so that a smooth abduction of the vocal cord would be achieved. In each animal, the chronically denervated muscle had a lower fusion frequency than its innervated partner. In a chronic implant, it would be desirable to lower the rate of stimulation under 30 Hz closer to that of the fusion frequency (mean: 21.77 Hz) to conserve charge. FIG. 3 shows views of a clinical patient with laryngeal hemiplegia both at rest and during stimulation with 4.5 mA at 24 Hz. As the pulse duration was increased, the efficiency in activating chronically denervated muscle increased and surpassed that of the innervated muscle at durations greater than 1-2 ms. However above 2 ms, stimulation became less efficient for both muscles because of charge loss through current shunts normally found in tissue. The amount of vocal cord excursion was only 40-70% of that produced with stimulation of the normally innervated muscle, indicative of denervation atrophy and loss of muscle contractility.
Sanders I et al., Arytenoid Motion Evoked By Regional Electrical Stimulation Of The Canine Posterior Cricoarytenoid Muscle, Laryngoscope. 1994 April; 104(4):456-62, incorporated herein by reference, systematically evaluated stimulation delivered to the denervated canine PCA muscles, using single-stranded, stainless steel electrodes 1 cm in length. Measures of abduction were obtained following an overdose of curare designed to mimic vocal fold paralysis via neuromuscular blockade. After RLN section and 2 weeks' time, measures of abduction were repeated in these animals. Results documented 3 mm of vocal cord excursion with 1 ms, 30 Hz, and 1-50 mA.
Sanders I., Electrical Stimulation Of Laryngeal Muscle, Otolaryngol Clin North Am. 1991 October; 24(5):1253-74, incorporated herein by reference, left 4 dogs undisturbed for 6 months to allow atrophy to occur. After 6 months of atrophy, the responses of the animals had decreased to roughly 60% of initial values. The two dogs that did not undergo stimulation continued to atrophy during the following 4 months to 40% of initial values. The two dogs that underwent electrically induced exercise, however, increased their responses dramatically. Not only had their responses returned to normal, but they were uniformly greater than normal, the average approximately 200% that of their initial denervated state. Gross examination of the excised larynges demonstrated that the stimulated group had maintained muscle bulk while the non-stimulated group was noticeably atrophic. Denervated dog PCA could be stimulated with pulses as short as 2 ms. Any lower, and the needed voltage jumped exponentially. Sanders used similar pulse widths to chronically stimulate denervated muscle for months. This is the minimum and presupposes that the electrode is placed directly adjacent to the muscle.
Zealear DL et al., Reanimation Of The Paralyzed Human Larynx With An Implantable Electrical Stimulation Device, Laryngoscope. 2003 July; 113(7):1149-56, incorporated herein by reference, reported on four human patients implanted with adapted pain pacemaker systems. In the four patients tested, electromyographic (EMG) motor unit activity was present in the PCA and thyroarytenoid (TA) muscles during voluntary effort. These recordings showed inappropriate firing patterns. For example, inspiratory motor unit activity was recorded from the TA muscle characteristic of a PCA motor unit. In particular, a deep inspiration or sniff increased the rate of firing of individual motor units and enhanced the overall interference response. This inappropriate activity was indicative of synkinetic reinnervation.
In follow-up sessions, the optimum stimulus parameters for vocal fold abduction were studied. A one- to two-second train of one-millisecond pulses delivered at a frequency of 30 to 40 pulses per second (pps) and amplitude of 2 to 7 V effectively produced a dynamic airway. One to two seconds of stimulated abduction allowed sufficient air exchange with each breath. Although a previous study in the canine found 2-millisecond duration as the optimum pulse width for recruiting both reinnervated and non-reinnervated muscle fibers, the maximum pulse width that the stimulator could deliver was 1 millisecond. A frequency of 30 to 40 pps generated a fused, tetanising muscle contraction and a smooth vocal fold abduction with maximum opening. The device was set to deliver an average of 10 stimulus sequences (bursts) every minute to match the patient's respiratory rate at a moderate level of activity. The ideal stimulus amplitude was one that evoked maximum vocal fold opening without inducing discomfort or nociception. At this amplitude, the patient could feel the stimulus, which helped entrain inspiration to the stimulus cycle. Stimulated abduction significantly increased the magnitude of glottal opening in patients 1 to 5 from preoperative levels (P<0.0008). Stimulated glottal opening was large in patients 1, 3, and 4 (3.5-7 mm) and moderate in patient 2 (3 mm). In patient 5, stimulation also produced a large abduction of 4 mm, but the response was delayed in time.
In order to decrease current spread and the high power requirements of FES devices, the placement of electrodes should localize current to the target muscle or nerve (if the muscle is innervated—even if it is synkinetically reinnervated) as much as possible. This may be accomplished by placing the electrodes inside the muscle, or on its surface, a procedure that produces two technical problems: (1) surgical exposure of the muscle causes scarring which eventually decreases muscle mobility; and (2) because electrodes must be close to their target to be efficient, they are exposed to muscle movement. The constant abrasion of the electrode against the muscle breaks the electrode or causes extensive fibrosis in the muscle. This difficulty plagued the early development of the cardiac pacer and persists today in many experiments involving chronic stimulation of denervated muscle, including the denervated PCA. As a result, there has not been a truly successful chronic device for stimulation of denervated muscle.
In 1992 for unilateral vocal cord paralysis, Goldfarb used the electric activity of the healthy side as a trigger for synchronization with breathing and vocalization. See, U.S. Pat. No. 5,111,814. This method is not applicable for the clinically more relevant bilateral paralysis. Lindenthaler described a pacemaker for bilateral vocal cord palsy due to autoparalysis (equivalent to synkinetic Recurrent Laryngeal Nerve (RLN) reinnervation), which is triggered by another muscle or nerve signal that is activated synchronic to breathing, e.g., diaphragm breathing muscles, infrahyoidal muscles of the neck. The pacemaker then stimulates structurally intact but autoparalytic nerve. See, U.S. Pat. No. 7,069,082.
For a real and complete rehabilitation of some patients with uni- or bilateral vocal cord paralysis or even in patients with a larynx transplantation a mere restoration of a single movement function of vocal cords by a pacemaker is not sufficient. In some cases even essential, is a pacemaker that can stimulate opening of vocal cords (e.g., to achieve sufficient breath for physical activities) as well as complete closure and tension of vocal cords (e.g., for vocalization and in combination with larynx elevation during swallowing for protection against aspiration). An optimal coordination of stimulated larynx movements with breathing cycle, intentional vocalization and swallowing reflex is necessary for that.
A stimulation of opening and closing of vocal cords might be helpful to preserve the full dynamic range of vocal cord movability by preventing a fixation of the cricoarytenoid joint. The necessary electrodes or sensing devices for detecting triggers and for stimulation of autoparalytic nerves or direct stimulation of paralyzed muscles of the larynx itself must not damage healthy tissue. In addition, the implantation procedure should also not cause harm.
Current surgical techniques all involve the exposure of the endings of the RLN or the exposure of the opening muscle (i.e., Posterior Cricoarytenoid Muscle, (PCA)) of the larynx. To achieve this, other muscles have to be cut (e.g., infrahyoidal muscles or pharyngeal constrictor muscle) and vessels and nerves in the vicinity may be damaged causing an impaired mobility of the larynx during swallowing and impaired sensitivity of mucus membranes with an increased risk of foreign body aspiration. Furthermore, scarring of all those tissues may diminish stimulated movements in the long run.
In addition, free placement of electrodes through the tissue to the target muscle (or nerve) may cause a high mechanical stress in the electrode leads which may cause lead wire breakage in delicate electrodes. Thus, placing or laying of the electrode in such a way that protects the electrode more may be helpful.
In order to decrease current spread and the high power requirements of FES devices, the placement of electrodes should localize current to the target muscle or nerve (if the muscle is innervated—even if it is synkinetically reinnervated) as much as possible. This may be accomplished by placing the electrodes inside the muscle or at its surface, a procedure that produces two technical problems: (1) surgical exposure of the muscle causes scarring which eventually decreases the muscle's mobility; and (2) because electrodes must be close to their target to be efficient, they are exposed to muscle movement. The constant abrasion of the electrode against muscle breaks the electrode or causes extensive fibrosis in the muscle. This difficulty plagued the early development of the cardiac pacer and persists today in many experiments involving chronic stimulation of denervated muscle, including the denervated PCA. As a result, there is not currently a truly successful chronic device for stimulation of denervated muscle.
An open surgery is much more invasive than a needle insertion. Insertion needles or puncture needles are typically straight and not curved, consisting of one part. For some situations, however, it is not possible to reach the target point (e.g., inside the subject's body or a different position outside the body then where the insertion started) in a straight line from the outside of the body or starting from cavities inside the body.